Over a decade of innovative collaborations among judicial authorities, drug treatment providers, social service agencies, researchers, and policy makers combine to present a first wave of support and enthusiasm for the drug court model with both adults and juveniles. Juvenile drug courts, like drug courts generally, have received encouraging evaluations of treatment retention, drug use, and recidivism during program participation, yet methodological limitations leave this early stage of work suspect. The current proposal, submitted in response to RFA DA-03-003, represents one of the first attempts to scientifically investigate the process and outcomes of juvenile drug court. The study addresses several gaps in the knowledge base on juvenile drug courts by examining the extent to which an empirically-supported treatment for adolescent drug abuse and delinquency, multidimensional family therapy (Liddle, 2002; MDFT), can enhance the effectiveness of existing juvenile drug court services in terms of increasing drug court graduation rates, decreasing drug use, delinquent behavior, and arrests, and improving school and vocational outcomes. The study design is a fully randomized controlled trial that compares the acceptability, efficacy, and benefit-cost of two different juvenile drug court models: MDFT integrated with existing core juvenile drug court services (MJDC-MDFT), and an individual and peer focused drug treatment intervention (Treatment as Usual; MJDC-TAU) integrated with existing core juvenile drug court services. In addition, we propose to investigate mechanisms of change by which both drug court models achieve their effects. The proposed study is responsive to the RFA in that it "builds on recent studies of drug abuse treatment for adolescents to improve and expand the delivery of efficacious treatments to drug abusing youth." Moreover, the proposed research will be conducted in collaboration with a project being considered by SAMHSA under its "Strengthening Communities-Youth" initiative (GFA TI 03-002). The proposed study has three aims: 1) To experimentally compare the acceptability and effectiveness of the Miami Juvenile Drug Court-Multidimensional Family Therapy (MJDC-MDFT) Program versus MJDC-TAU; 2) To identify the mechanisms through which the drug court programs achieve their effects; and 3) To compare the total and net monetary benefits to the juvenile justice system of serving juvenile drug court participants with MJDC-MDFT vs. MJDC-TAU. A total of 150 juveniles eligible for the Miami Juvenile Drug Court and their parents will comprise the study sample (75 per condition). Eligible youth will be randomly assigned to either MJDC-MDFT or MJDC-TAU and will be assessed at baseline, 6 months, 12 months, 18 months, and 24 months following baseline to the drug court program. State-of-the-art statistical techniques, including latent growth curve modeling (LGM), will be used to examine change in multiple processes and outcomes over time.